SFDA-rPPG: Source-Free Domain Adaptive Remote Physiological Measurement With Spatiotemporal Consistency

Remote photoplethysmography (rPPG) is a noncontact method that uses facial video to predict changes in blood volume, enabling physiological metrics. Traditional rPPG models often struggle with poor generalization capacity in unseen domains. Current solutions to this problem are to improve its generalization in the target domain through domain generalization (DG) or domain adaptation (DA) techniques. However, traditional DA methods usually require access to both source-domain data and target-domain data, which cannot be implemented in scenarios with limited access to source data due to privacy issues. In this article, we propose the first source-free DA benchmark for rPPG (SFDA-rPPG) measurement, which overcomes these limitations by enabling effective DA without access to source-domain data. Our framework incorporates a three-branch spatiotemporal consistency network (TSTC-Net) to enhance feature consistency across domains. Furthermore, we propose a new rPPG distribution alignment loss based on the frequency-domain Wasserstein distance (FWD), which leverages optimal transport to align power spectrum distributions across domains effectively and further enforces the alignment of the three branches. Extensive cross-domain experiments and ablation studies demonstrate the effectiveness of our proposed method in source-free DA (SFDA) settings. Our findings highlight the significant contribution of the proposed FWD loss for distributional alignment, providing a valuable reference for future research and applications.